One of your problems, if you so choose to add some sort of ties each side of the existing bottom flange, will be connecting to the ends of the existing truss to resist and be equal to the compressive from the top chords. The truss should now be shored before you begin any work on it. A procedure might be shoring the truss - maybe in three places and jacking the center some distance above the design bottom chord elevation; installing the new tie rods and anchorage and then removing the shoring. You'll have to calculate the theoritical difference of the deadload deflection with the new configuration.
If there's problems with these trusses, I would shore all of them until the the repairs are completed. One truss failure will probably bring all of them down. (Investigated similar situation in Richmond in the early 90's where one truss failed and because of piping and other junk attached over the years, brought all of the rest of the trusses down. (about 30!)
Interesting item was that there had been a number of repairs over the years of broken webs and split connections (upper and lower) chords that were mickey mouse. Ovisously never reviewed by an engineer or a city inspector. (Trusses installed in the 1950's during the Korean War). During the failure there were six people in the building but because the building stored wreck cars that were being parted out, the trusses came down and rested on top of the cars and nobody was hurt. An argument ensued in the office as to which truss failed first and caused the ensuing ripple effect. I maintained that a certain truss had failed and was challenged how I could state this. After some discussion (understatement), I admitted that I had spoken with an eye witness. Frankly I had no idea which truss failed and there were a number of candidates.
